Research Article
DEDGCN: Dual Evolving Dynamic Graph Convolutional Network
Table 1
A summary of dynamic network embedding methods.
| | Method | Learning techniques | Supervised | Unsupervised |
| | DHPE [16] | Matrix decomposition, embedded update | | √ | | TIMERS [17] | Matrix decomposition, embedded update | | √ | | DyREP [18] | Dynamic network structure characteristics | | √ | | DynamicTriad [19] | Dynamic network structure characteristics | | √ | | GCRN [12] | Splicing GCN and RNN | √ | | | WD-GCN/CD-GCN [20] | Splicing GCN and RNN | √ | | | RgCNN [13] | Splicing GCN and RNN | √ | | | Addgraph [21] | Attentional mechanism evolves GCN | √ | | | EvolveGCN [14] | RNN evolves GCN | √ | | | DynGEM [22] | Scalable autoencoder network | | √ | | Dyngraph2vec [23] | LSTM evolves autoencoder network | | √ |
|
|
